Lignin is one of the major products of the general phenylpropanoid pathway, and is one of the most abundant organic molecules in the biosphere (Crawford, (1981) Lignin Biodegradation and Transformation, New York: John Wiley and Sons). In nature, lignification provides rigidity to wood and is in large part responsible for the structural integrity of plant tracheary elements. Lignin is well suited to these capacities because of its physical characteristics and its resistance to biochemical degradation. Unfortunately, this same resistance to degradation has a significant impact on the utilization of lignocellulosic plant material (Whetten et al., Forest Ecol. Management 43, 301, (1991)).
The monomeric composition of lignin has significant effects on its chemical degradation during industrial pulping (Chiang et al., Tappi, 71, 173, (1988). The guaiacyl lignins (derived from ferulic acid) characteristic of softwoods such as pine, require substantially more alkali and longer incubations during pulping in comparison to the guaiacyl-syringyl lignins (derived from ferulic acid and sinapic acid) found in hardwoods such as oak. The reasons for the differences between these two lignin types has been explored by measuring the degradation of model compounds such as guaiacylglycerol-.beta.-guaiacyl ether, syringylglycerol-.beta.-guaiacyl ether, and syringylglycerol-.beta.-(4-methylsyringyl) ether (Kondo et al., Holzforschung, 41, 83, (1987)) under conditions that mimic those used in the pulping process. In these experiments, the mono- and especially di-syringyl compounds were cleaved three to fifteen times faster than their corresponding diguaiacyl homologues. These model studies are in agreement with studies comparing the pulping of Douglas fir and sweetgum wood where the major differences in the rate of pulping occurred above 150.degree. C. where arylglycerol-.beta.-aryl ether linkages were cleaved (Chiang et al., Holzforschung, 44, 309, (1990)).
Another factor affecting chemical degradation of the two lignin forms may be the condensation of lignin-derived guaiacyl and syringyl residues to form diphenylmethane units. The presence of syringyl residues in hardwood lignins leads to the formation of syringyl-containing diphenylmethane derivatives that remain soluble during pulping, while the diphenylmethane units produced during softwood pulping are alkali-insoluble and thus remain associated with the cellulosic products (Chiang et al., Holzforschung, 44, 147, (1990); Chiang et al., Holzforschung, 44, 309, (1990)). Further, it is thought that the abundance of 5-5'-diaryl crosslinks that can occur between guaiacyl residues contributes to resistance to chemical degradation. This linkage is resistant to alkali cleavage and is much less common in lignin that is rich in syringyl residues because of the presence of the 5-O-methyl group in syringyl residues. The incorporation of syringyl residues results in what is known as "non-condensed lignin", a material that is significantly easier to pulp than condensed lignin.
Similarly, lignin composition and content in grasses is a major factor in determining the digestibility of lignocellulosic materials that are fed to livestock (Jung, H. G. & Deetz, D. A. (1993) Cell wall lignification and degradability in Forage Cell Wall Structure and Digestibility (H. G. Jung, D. R. Buxton, R. D. Hatfield, and J. Ralph eds.), ASA/CSSA/SSSA Press, Madison, Wis.). The incorporation of the lignin polymer into the plant cell wall prevents microbial enzymes from having access to the cell wall polysaccharides that make up the wall. As a result, these polysaccharides cannot be degraded and much of the valuable carbohydrates contained within animal feedstocks pass through the animals undigested. Thus, an increase in the dry matter of grasses over the growing season is counteracted by a decrease in digestibility caused principally by increased cell wall lignification. From these examples, it is clear that the modification of lignin monomer composition would be economically advantageous.
The problem to be overcome, therefore, is to develop a method for the creation of plants with increased levels of syringyl residues in their lignin to facilitate its chemical degradation. Modification of the enzyme pathway responsible for the production of lignin monomers provides one possible route to solving this problem.
The mechanism(s) by which plants control lignin monomer composition has been the subject of much speculation. As mentioned earlier, gymnosperms do not synthesize appreciable amounts of syringyl lignin. In angiosperms, syringyl lignin deposition is developmentally regulated: primary xylem contains guaiacyl lignin, while the lignin of secondary xylem and sclerenchyma is guaiacyl-syringyl lignin (Venverloo, Holzforschung 25, 18 (1971); Chapple et al., Plant Cell 4, 1413, (1992)). No plants have been found to contain purely syringyl lignin. It is still not clear how this specificity is controlled; however, at least five possible enzymatic control sites exist, namely caffeic acid/5-hydroxyferulic acid O-methyltransferase (OMT), F5H, (hydroxy)cinnamoyl-CoA ligase (4CL), (hydroxy)cinnamoyl-CoA reductase (CCR), and (hydroxy)cinnamoyl alcohol dehydrogenase (CAD). For example, the substrate specificities of OMT (Shimada et al., Phytochemistry, 22, 2657, (1972); Shimada et al., Phytochemistry, 12, 2873, (1973); Gowri et al., Plant Physiol, 97, 7, (1991); Bugos et al., Plant Mol. Biol. 17, 1203, (1992)) and CAD (Sarni et al., Eur. J Biochem., 139, 259, (1984); Goffner et al., Planta., 188, 48, (1992); O'Malley et al., Plant Physiol., 98, 1364, (1992)) are correlated with the differences in lignin monomer composition seen in gymnosperms and angiosperms, and the expression of 4CL isozymes (Grand et al., Physiol. Veg. 17, 433, (1979); Grand et al., Planta., 158, 225, (1983)) has been suggested to be related to the tissue specificity of lignin monomer composition seen in angiosperms.
Although there are at least five possible enzyme targets that could be exploited, only OMT and CAD have been investigated in recent attempts to manipulate lignin monomer composition in transgenic plants (Dwivedi et al., Plant Mol. Biol. 26, 61, (1994); Halpin et al., Plant J 6, 339, (1994); Ni et al., Transgen. Res. 3, 120 (1994); Atanassova et al., Plant J. 8, 465, (1995); Doorsselaere et al., Plant J 8, 855, (1995)). Most of these studies have focused on sense and antisense suppression of OMT expression. This approach has met with variable results, probably owing to the degree of OMT suppression achieved in the various studies. The most dramatic effects were seen by using homologous OMT constructs to suppress OMT expression in tobacco (Atanassova et al., supra) and poplar (Doorsselaere et al., supra). Both of these studies found that as a result of transgene expression, there was a decrease in the content of syringyl lignin and a concomitant appearance of 5-hydroxyguaiacyl residues. As a result of these studies, Doorsselaere et al., (WO 9305160) disclose a method for the regulation of lignin biosynthesis through the genomic incorporation of an OMT gene in either the sense or anti-sense orientation. In contrast, Dixon et al. (WO 9423044) demonstrate the reduction of lignin content in plants transformed with an OMT gene, rather than a change in lignin monomer composition. Similar research has focused on the suppression of CAD expression. The conversion of coniferaldehyde and sinapaldehyde to their corresponding alcohols in transgenic tobacco plants has been modified with the incorporation of an A. cordata CAD gene in anti-sense orientation (Hibino et al., Biosci. Biotechnol. Biochem., 59, 929, (1995)). A similar effort aimed at antisense inhibition of CAD expression generated a lignin with increased aldehyde content, but only a modest change in lignin monomer composition (Halpin et al., supra). This research has resulted in the disclosure of methods for the reduction of CAD activity using sense and anti-sense expression of a cloned CAD gene to effect inhibition of endogenous CAD expression in tobacco [Boudet et al., (U.S. Pat. No. 5,451,514) and Walter et al., (WO 9324638); Bridges et al., (CA 2005597)]. None of these strategies increased the syringyl content of lignin, a trait that is correlated with improved digestibility and chemical degradability of lignocellulosic material (Chiang et al., supra; Chiang and Funaoka, Holzforschung 44, 309 (1990); Jung et al., supra).
Although F5H is also a key enzyme in the biosynthesis of syringyl lignin monomers it has not been exploited to date in efforts to engineer lignin quality. In fact, since the time of its discovery over 30 years ago (Higuchi et al., Can. J Biochem. Physiol., 41, 613, (1963)) there has been only one demonstration of the activity of F5H published (Grand, C., FEBS Lett. 169, 7, (1984)). Grand demonstrated that F5H from poplar was a cytochrome P450-dependent monooxygenase (P450) as analyzed by the classical criteria of dependence on NADPH and light-reversible inhibition by carbon monoxide. Grand further demonstrated that F5H is associated with the endoplasmic reticulum of the cell. The lack of attention given to F5H in recent years may be attributed in general to the difficulties associated with dealing with membrane-bound enzymes, and specifically to the lability of F5H when treated with the detergents necessary for solubilization (Grand, supra). The most recent discovery surrounding the F5H gene has been made by Chapple et al., (supra) who reported a mutant of Arabidopsis thaliana L. Heynh named fah1 that is deficient in the accumulation of sinapic acid-derived metabolites, including the guaiacyl-syringyl lignin typical of angiosperms. This locus, termed FAH1, encodes F5H. The cloning of the gene encoding F5H would provide the opportunity to test the hypothesis that F5H is a useful target for the engineering of lignin monomer composition.
In spite of sparse information about F5H in the published literature, Applicant has been successful in the isolation, cloning, and sequencing of the F5H gene. Applicant has also demonstrated that the stable integration of the F5H gene into the plant genome, where the expression of the F5H gene is under the control of a promoter other than the gene's endogenous promoter, leads to an altered regulation of lignin biosynthesis.